Book Image

Hands-On Data Science with R

By : Vitor Bianchi Lanzetta, Doug Ortiz, Nataraj Dasgupta, Ricardo Anjoleto Farias
Book Image

Hands-On Data Science with R

By: Vitor Bianchi Lanzetta, Doug Ortiz, Nataraj Dasgupta, Ricardo Anjoleto Farias

Overview of this book

R is the most widely used programming language, and when used in association with data science, this powerful combination will solve the complexities involved with unstructured datasets in the real world. This book covers the entire data science ecosystem for aspiring data scientists, right from zero to a level where you are confident enough to get hands-on with real-world data science problems. The book starts with an introduction to data science and introduces readers to popular R libraries for executing data science routine tasks. This book covers all the important processes in data science such as data gathering, cleaning data, and then uncovering patterns from it. You will explore algorithms such as machine learning algorithms, predictive analytical models, and finally deep learning algorithms. You will learn to run the most powerful visualization packages available in R so as to ensure that you can easily derive insights from your data. Towards the end, you will also learn how to integrate R with Spark and Hadoop and perform large-scale data analytics without much complexity.
Table of Contents (16 chapters)

Neural networks

These are the models that most fascinate me. From their wide range of applications to their brilliant origin—everything seems marvelous to me. Note that there is no such a thing as an all mighty model and neural nets are not it. They can be distinguished by being very flexible models that can perform both unsupervised and supervised learning.

Even though it's a very powerful method, it's certainly not all powerful. Neural nets are able to capture linear and non-linear relations. Yet, everything will depend on how you design the networks (researcher's ability and experience), how complex is the problem at hand, and how many observations do you have.

Computational constraints can also be a bottleneck. As powerful as they are known to be, neural networks are not remembered as computationally inexpensive methods. It's not hard to find problems...